> Am 16.02.2016 um 09:57 schrieb William Hay <w....@ucl.ac.uk>: > > I'm getting messages like the following popping up in the grid engine > messages file (SoGE 8.1.8): > 02/16/2016 08:20:07|worker|nfs-2|E|invalid task number 0 for job 496841 in > "ORT_ptickets" order
I never saw it before. And it's not an array task which got an illegal lower boundary? -- Reuti > 02/16/2016 08:20:07|worker|nfs-2|W|Skipping remaining 352 orders > 02/16/2016 08:20:07|worker|nfs-2|E|reinitialization of "scheduler" > > They persist over scheduling cycles but can generally be cleared by applying > a qhold on the job in question followed > by a qrls. > > They seem[1] to prevent scheduling for subsequent jobs and quite often once > I've cleared one job another job submitted > by the same user at about the same time will start showing up in the logs > with the same messages. > > Has anyone seen something similar and/or know what causes it? > > [1]'Seem' as in this is happening on our production cluster and my priority > has been to clear the blockage rather > than investigate in detail. > _______________________________________________ > users mailing list > users@gridengine.org > https://gridengine.org/mailman/listinfo/users
signature.asc
Description: Message signed with OpenPGP using GPGMail
_______________________________________________ users mailing list users@gridengine.org https://gridengine.org/mailman/listinfo/users
